Various conventional content delivery systems enable the user of a computing device (e.g., a smartphone) to obtain content related to an image the user has provided, such as by capturing the image using a camera of the computing device. For example, a user may capture an image of a movie poster through the camera and upload the captured image to a content delivery system. In turn, the content delivery system returns content related to the movie associated with the movie poster (e.g., information about the movie, multimedia content, etc.).
In order to deliver content related to the uploaded image, a conventional content delivery system attempts to match the image received from the client device against images stored in the content delivery system's image database. The image database can store thousands of images from magazines, posters, newspapers, the Internet, etc. For example, the image database can store the images of wall posters and billboard advertisements corresponding to a particular movie “X”. Consequently, when the content delivery system receives the image of the poster of the movie “X” from a communication device, the system attempts to match the received image with all the images stored in its image database. Such an approach can be very resource intensive, and can provide significant lag time when a user submits a request to a system with a large set of images that must be analyzed.